Skip to content

Instantly share code, notes, and snippets.

@lfy79001
Last active April 22, 2024 12:10
Show Gist options
  • Save lfy79001/0e3eaf361ed11a1ae8b5f8ddb92ed6e8 to your computer and use it in GitHub Desktop.
Save lfy79001/0e3eaf361ed11a1ae8b5f8ddb92ed6e8 to your computer and use it in GitHub Desktop.
80b981ba-be6a-48dd-b4c6-180331a668a9
WITH base_table AS (
-- pulls relevant columns from relevant dates to decrease the size of data scanned
SELECT
event_name,
event_date,
event_timestamp,
user_pseudo_id,
user_id,
device,
geo,
traffic_source,
event_params,
user_properties
FROM
`bigquery-public-data.ga4_obfuscated_sample_ecommerce.events_*`
WHERE
_table_suffix = '20210101'
AND event_name IN ('page_view')
)
, unnested_events AS (
-- unnests event parameters to get to relevant keys and values
SELECT
event_date AS date,
event_timestamp AS event_timestamp_microseconds,
user_pseudo_id,
MAX(CASE WHEN c.key = 'ga_session_id' THEN c.value.int_value END) AS visitID,
MAX(CASE WHEN c.key = 'ga_session_number' THEN c.value.int_value END) AS visitNumber,
MAX(CASE WHEN c.key = 'page_title' THEN c.value.string_value END) AS page_title,
MAX(CASE WHEN c.key = 'page_location' THEN c.value.string_value END) AS page_location
FROM
base_table,
UNNEST (event_params) c
GROUP BY 1,2,3
)
,unnested_events_categorised AS (
-- categorizing Page Titles into PDPs and PLPs
SELECT
*,
CASE WHEN ARRAY_LENGTH(SPLIT(page_location, '/')) >= 5
AND
CONTAINS_SUBSTR(ARRAY_REVERSE(SPLIT(page_location, '/'))[SAFE_OFFSET(0)], '+')
AND (LOWER(SPLIT(page_location, '/')[SAFE_OFFSET(4)]) IN
('accessories','apparel','brands','campus+collection','drinkware',
'electronics','google+redesign',
'lifestyle','nest','new+2015+logo','notebooks+journals',
'office','shop+by+brand','small+goods','stationery','wearables'
)
OR
LOWER(SPLIT(page_location, '/')[SAFE_OFFSET(3)]) IN
('accessories','apparel','brands','campus+collection','drinkware',
'electronics','google+redesign',
'lifestyle','nest','new+2015+logo','notebooks+journals',
'office','shop+by+brand','small+goods','stationery','wearables'
)
)
THEN 'PDP'
WHEN NOT(CONTAINS_SUBSTR(ARRAY_REVERSE(SPLIT(page_location, '/'))[SAFE_OFFSET(0)], '+'))
AND (LOWER(SPLIT(page_location, '/')[SAFE_OFFSET(4)]) IN
('accessories','apparel','brands','campus+collection','drinkware',
'electronics','google+redesign',
'lifestyle','nest','new+2015+logo','notebooks+journals',
'office','shop+by+brand','small+goods','stationery','wearables'
)
OR
LOWER(SPLIT(page_location, '/')[SAFE_OFFSET(3)]) IN
('accessories','apparel','brands','campus+collection','drinkware',
'electronics','google+redesign',
'lifestyle','nest','new+2015+logo','notebooks+journals',
'office','shop+by+brand','small+goods','stationery','wearables'
)
)
THEN 'PLP'
ELSE page_title
END AS page_title_adjusted
FROM
unnested_events
)
SELECT * FROM unnested_events_categorised;
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment